Category: ADHD - Adult
With only recent recognition of the persistence of ADHD into adulthood (Wolraich et al., 2012), best practice in identifying and diagnosing ADHD in adults is unconfirmed (Sibley et al., 2012). Researchers have examined tools for evaluating ADHD symptoms in adults (Sibley et al., 2016), however, tools for identifying related impairment are lacking. The Impairment Rating Scale (IRS; Fabiano et al., 2006), which was originally designed for children and modified to be applicable for adults, functions as a screener of multi-domain impairment. The IRS is used to help identify adults with ADHD (e.g., Sibley et al., 2016). However, the psychometric properties and appropriate cut-scores for clinically significant impairment have not been evaluated with adults. The current study assessed the psychometric properties of the IRS across three samples and sought to establish an empirically informed cut-off for diagnostic and research purposes for adults with ADHD.
Three samples over-recruited for adults with ADHD included a group of college students (N=752; 38.6% male; 88.0% white; M age=19.14 years), an online population (N=346; 55.5% male; 81.8% white; M age=27.99), and emerging adult couples (N=61 couples; 50% male; 92.4% white; M age=20.32). IRS items are rated on a continuum from “No Problem” to “Extreme Problem,” recoded numerically (0-6). Convergent validity was evaluated by determining associations with areas of impairment for adults with ADHD [e.g., academic grades, substance use, intimate partner violence]. In one sample, inter-rater reliability between partners was used to examine the stability of the IRS. Finally, sensitivity, specificity, positive predictive values (PPV), and negative predictive values (NPV) were examined at various cut-points to identify a justifiable IRS cut-point to help determine ADHD diagnostic status.
Across all samples, IRS items were associated with other measures of impairment indicative of convergent validity (e.g., greater romantic relationship impairment was positively associated with intimate partner violence). Evidence of discriminant validity (e.g., IRS items unrelated to religion and age) was observed across samples and stability across raters within couples was observed in one sample. Finally, upon examination of sensitivity, specificity, PPV, and NPV, a score of 1+ on the IRS emerged as an appropriate cutoff to aid in the diagnostic evaluation of individuals with ADHD. These findings support the use of the IRS for assessing ADHD-related impairment in adult populations, and provide a first step in establishing a cut-point for clinically significant impairment. Implications for use of the IRS as a diagnostic tool within adult populations, as well as future directions will be discussed.